Hematopoietic cytokines stimulate proliferation and survival, but their role in lineage commitment remains uncertain. To gain insight into the contribution of cytokine signaling to myeloid lineage specification, we compared G-CSF and M-CSF signaling in primary murine, lineage-negative marrow cells. G-CSF specifically activated STAT3 and induced increased SHP2 phosphorylation, whereas M-CSF preferentially activated PLC-γ and ERK and induced c-Fos in marrow progenitors. In contrast, phosphorylation of Jun kinase or c-Jun was similar in response to both cytokines, as was induction of PU.1. Inhibition of ERK by either of two chemical inhibitors prevented c-Fos induction and interfered with monocyte lineage specification, as assessed by enumerating monocytic versus granulocytic colony-forming units in methylcellulose culture with IL3/IL6/SCF. ERK inhibition also reduced C/EBPα21 phosphorylation, as seen previously in cell lines. M-CSF induced C/EBPαS21 modification more effectively than G-CSF, but the effect of this modification on C/EBPα activities and its role in myeloid lineage choice remains uncertain. Consistent with a role for ERK in monocyte specification, phorbol ester potently activated ERK and induced monocytic colonies in IL3/IL6/SCF or in GCSF/SCF methylcellulose culture. SHP2 inhibition reduced ERK activation by G-CSF but not by M-CSF and reduced granulopoiesis but not monopoiesis in CFU assays with IL3/IL6/SCF, indicating that an ERK-independent target of SHP2 helps specify granulopoiesis. Phorbol ester induction of ERK and monopoiesis may reflect preferential activation of PLC-γ by M-CSF compared to G-CSF, as we observed. In summary, M-CSF signals activate ERK more potently than G-CSF signals and thereby induce c-Fos to favor monopoiesis, whereas G-CSF signals activate the SHP2 tyrosine phosphatase, shifting the balance to granulopoiesis, via effects of SHP2 on additional regulators besides ERK. How cytokine signals cooperate with transcription factors to specify lineage determination is a fundamental issue in hematopoiesis, and elucidating these functional interactions will facilitate efforts to expand normal myeloid cells and to use cytokines or downstream effectors to induce malignant cells to differentiate.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author